Originally Posted by
Tomaski
but the two issues are really inseparable.
maybe
maybe not
An aircraft stalling accidentally isn't due to manual flying skills, it is because the driver doesn't recognise his energy or spatial orientation state. Now driving holes around the pattern will be both fun and assist to develop some general awareness, but, if you call for me to set flaps 30 and instead I select flap 5 from 15, does your hand flying skills actually come into play or not. There is a trim change, and an attitude change, but as the hand flyer is effectively pushing and shoving to make needles point to where they are eye pleasing, the pilot won't necessarily pick up the slip between expectation and reality. [maybe 5-10% of pilots will detect an anomaly through proprioceptive triggers, some more will crosscheck the setting of the flap vs commanded, and some of those will actually ping to the fact that the wrong selection has been made... the rest will be oblivious to the difference from required thrust, attitude etc that is needed to fly at the erroneous configuration]. Or, you are taking off out of KLAX on 07L for a change of scenery, and ATC gives you a left turn to 180 instead of a right turn to 180. As you are an english speaker, convert the languages to this being Kunming, same scenario... ATC has made an error. Your manipulative skills don't assist directly in avoiding the turf. Say you are instead taking off out of Sharm Al Sheik, doing a turn out over the darkness, and your outer slat does not retract when you select flaps 0. Does manual flying skills actually alter the outcome in that instance, where the uncommanded roll occurs but is within the lateral authority of the flight controls. The slip between the real world and the expected world is what gives the bad hair day IMHO. Same story for say a B747 in the cruise and having an outboard engine failure and the crew dithering about a response, while the aircraft is above the driftdown height. In all of these cases, the crews perception of the state of the aircraft aerodynamically, geographically, or system wise is not the same as the world as it is (or was) and that needs to be resolved. The hand flying activity is nice to have, it improves the confidence of the flight crew and their ability to toss away automation and revert when said automation is doing undesirable antics, but not really that much beyond.
It is possible to train for SA, the US Army has been doing that for a while, in ascertaining an individual or teams level of SA, when it becomes compromised, training for recognition of loss of SA, and training for recovery of SA. the interesting thing is it doesn't take a 6-DOF device to do that. SA is the "head-fake" (1) we play on ourselves. Every pilot who ever parked an aircraft into the side of a mountain was surprised to see the mountain goats. A lot of those planes that got parked were also being driven by handraulics at the time of impact A disproportionate number of handflown aircraft have parked unceremoniously outside of the MM.
I had the chance to go through a major airlines rather considerable data set once upon a time, which included a large number of wadded aircraft, from small to top of the line, and also a comprehensive set of serious events, where generally the aircraft remained useable after a fashion. The data set had a population of flight data rather larger than that necessary to get a normal distribution for any particular characteristic, think of a number between 3000 and 5000 events. (it was a big program, and, well they had their challenges that they were attempting to meet). Two points stood out of the data. Actually about 4 things did.
- events that were crew involved were usually not reported to the company, they were detected. That is about the first point of interest in determining if y'all may have a pathology in your management process.
- crew reports when made were 50% ATC, 40% BIRD (or fish, said mountain goats if the crew had survived...) and 90% MECHANICAL. Yep, they really didn't add up to making any sense at all... excuse the literary license...
- 98% ( an actual statistic found by analysis of each events data) of all serious events had a point which could be identified where SA was lost. The company analysis previously had not evaluated their data set in such a manner...
- The crews were reticent in reverting to manual modes, even when stuff was getting weird. The aforementioned management had a policy that added some latency in the decision making of the drivers, there was a vice like grip on the trivial, and infractions were effectively binary conditions, not a fuzzy logic analysis of soft edges.
Hand driving may assist in SA, but it is not a direct outcome it is osmotic at best. what a pilot does with their hands has a weak linkage to their awareness of their operational state, it may actually reduce that on occasions through the limited capability of multitasking that humans have. (We are essentially serial processors, not truly parallel, other than very well established manipulative processes. The more trained we are the closer it looks to being parallel processing, as the access time to gain inputs to correct a control requirement becomes more refined. We employ scanning effectively as a technique to maintain a reasonable control of multivariate inputs, but add a cross check or additional cognitive load and we skip a beat. If I call as a support pilot Gear, checked, 3 green, no red, I can almost guarantee that a pilot handflying the approach with any dynamic demands will parrot back the same response, without looking... and even if they do, they as often as not will still say what they heard, not what they see).
What does all this suggest? Merely that hand flying itself is not the panacea to the problems that we see in our profession, it is a nice thing to have, and has some use but it is as much as anything a nostalgic look back to the future. When we didn't have good systems we yearned for them, when we have them, the fact that we are still human has us looking backward to the good old days, of aircraft that had awful ergonomics, lousy radar, inaccurate and unreliable nav systems, lousy approaches. smoky cockpits... etc. The engines used to cough and wheeze, stall and surge, and otherwise make life annoying. The lack of fuel capacity and range gave us interesting tech stops and overnights. in places that are on travel alert lists frequently. Ah, the good old days. Where trees and boundary fences were always bigger than expected on departures. And notwithstanding the fact that we hand flew the planes routinely, we buried planes in every tall hill and every ocean, parked short of runways, overran the other end to balance the mess about the concrete. Was our SA better for it? Not sure that it was.
keeping on doing the same experiment expecting the outcome to change is the basis of one of Einstein's quips. Another was:
"A man should look for what it is, and not for what should be..."
(1) the last lecture of Randy Pausch, PhD; September 18, 2007, Carnegie Mellon Uni. Pittsburgh, PA